The physiology of intercellular signaling through gap junctions is still a mystery. In spite of considerable progress in understanding the biochemistry, genetics and biosynthesis of junctional channels, the character of their most salient functional property - permeability to signaling molecules - remains unknown. Which chemical signals go through connexin channels, and how well, are fundamental issues with far-reaching impact. Gap junction channels (composed of connexin) form regulated pathways mediating direct intercellular transfer of ions and small molecules. Study of the pathway is constrained by its location in situ; both ends of the pore are intracellular, inaccessible to most manipulations that explore channel selectivity. Studies of connexin permeation tend to focus on small atomic ions or large fluorescent tracers. Yet, it is the selectivity among specific signaling molecules that is of key biological importance. Also, since access is via cytoplasm, it is difficult to identify modulatory factors acting directly on the channel to alter permeability, rather than via cellular components. The long-term objective is to understand the molecular operation of this pathway of intercellular communication. The approach is to study connexin channels in a reconstituted system where their selective permeation properties can be fully explored. This study uses channels formed by connexin32 and connexin26 immunopurifield from native tissues and expression vectors in a well-characterized system that yields information not available from cellular studies. The proposed experiments address these questions: Do connexin channels select among signaling molecules (e.g., second messengers) by specific molecular affinities? What are the distinctive molecular selectivities of the different connexins? How do molecular size and charge affect what can permeate the channels? What are the direct modulators of connexin channels, and their mechanisms of action? By study of connexin channels in an experimentally accessible system, one hopes to understand the fundamental properties of junctional communication. Gap junctions are so widespread that elucidation of connexin channel selectivity will have profound consequences throughout cellular and developmental biology. There are currently 12 known connexins. Genetic defects in connexin32 cause a peripheral neuropathy, and in connexin43 defects of cardiac development. No doubt many other syndromes arise in toto or in part from defects in connexin channel function. A functional ~defect~ of connexin channels will be reflected in an abnormal intercellular permeability (high or low) to a cytoplasmic molecule. The proposed studies address the basis for how this may occur.